The Dow Chemical Company announced [Me2Si(Me4C5)NtBu]TiCl2 (Constrained-Geometry Catalyst, which will be simply referred to as CGC) in the early 1990's (U.S. Pat. No. 5,064,802), wherein in the copolymerization reaction of ethylene and alpha-olefin, excellent characteristics that the CGC has, as compared with per se known metallocene catalysts can be usually classified into the two categories: (1) it produces a high molecular weight polymer with high activity even at a high polymerization temperature, and (2) it yield very excellent copolymerization of an alpha-olefin having high steric hindrance, such as 1-hexene and 1-octene. In addition, upon polymerization reaction, there have been gradually several characteristics of CGC, and thus extensive studies to synthesize a derivative of CGC for use as a polymerization catalyst have been made in the academic and industrial fields.
As one approach, there have been trials for synthesis of metal compounds to which various bridges and nitrogen substituents instead of silicon bridges are introduced, and polymerization using the same. Some representative examples of recently known metal compounds include the followings (Chem. Rev. 2003, 103, 283):

To the above-listed compounds, a phosphorous bridge (1), an ethylene or propylene bridge (2), a methylidene bridge (3), and a methylene bridge (4) are each introduced, instead of the silicon bridges in the CGC structure. However, when they are used for ethylene polymerization, or copolymerization with an alpha-olefin, they had no excellent results regarding the activity or the copolymerization performances, as compared with CGC.
As another approach, there have trials for synthesis of many compounds comprising an oxido ligand instead of the amido ligand of the CGC, and sometimes polymerization using the same. Examples thereof are summarized as follows:

The compound (5) is characterized in that a Cp derivative and an oxido ligand are bridged via an orthophenylene group, as disclosed by T. J. Marks, et al. (Organometallics 1997, 16, 5958). Also, a compound having the same bridge and polymerization using the same are disclosed by Mu, et al. (Organometallics 2004, 23, 540). Further, an indenyl ligand and an oxido ligand are bridged via the same orthophenylene group, as disclosed by Rothwell, et al. (Chem. Commun. 2003, 1034). The (6) is characterized in that a cyclopentadienyl ligand and an oxido ligand are bridged through three carbons, as disclosed by Whitby, et al. (Organometallics 1999, 18, 348), and these catalysts are reported to exhibit activity on syndiotactic polystyrene polymerization. Further, similar compounds are also reported by Hessen, et al. (Organometallics 1998, 17, 1652). The compounds shown in (7) is characterized in that it exhibits activity on ethylene and ethylene/1-hexene copolymerization at a high temperature and a high pressure (210° C., 150 MPa), as disclosed by Rau, et al. (J. Organomet. Chem. 2000, 608, 71). Further, synthesis of a catalyst having the similar structure (8), and polymerization using the catalyst at a high temperature and a high pressure were filed in the patent application by Sumitomo Corp. (U.S. Pat. No. 6,548,686).
Recently, a compound having two metal positions was developed, and has been employed as a polymerization catalyst since then. U.S. Pat. No. 6,153,776 describes a binuclear metal compound in which two compounds in the CGC form are bridged via a cyclopentadienyl (Cp) group, an Si bridge, or an N substituent group, and a method for preparing an olefin polymer using the same. Marks, et al., [J. Am. Chem. Soc., 2004, 126, 6542-6543] describes that if a binuclear Ti metal compound in which two compounds in the CGC form are bridged via a —CH2CH2— group is used for styrene homopolymerization and ethylene/styrene copolymerization reactions, the polymerization activity is more excellent than that of the conventionally used CGC compounds. Further, Marks, et al, describes an example in which a Ti compound and a Zr compound, each in the CGC form, are sequentially bridged to prepare a binuclear Ti/Zr metal compound, and this compound can be used for an ethylene homopolymerization reaction for preparation of polyethylene having a branch, even without comonomers. Accordingly, the above prior arts show that by using a binuclear metal compound for an olefin polymerization reaction, a polyolefin having more excellent activity and novel physical properties can be prepared, as compared with when using the conventionally used single metal compound.
However, in spite of the above trials, only few catalysts are substantially in use for commercial plants. Thus, there is still a need of development of a catalyst having a novel structure, and preparation of a polymer using the same. Particularly, the conventional binuclear catalyst compounds still have structures in the CGC form, and the method for preparing the same is complex and the production yield is low. Therefore, there is still a desire of a binuclear transition metal compound having a novel structure, which is capable of exhibiting a catalytic activity, and a method for simply preparing the same in high yield.